Clinician-Directed Large Language Model Software Generation for Therapeutic Interventions in Physical Rehabilitation
🤖 AI Summary
Current digital rehabilitation interventions rely on static, pre-defined templates, resulting in poor clinical adaptability and limited responsiveness to individual patient needs—such as physical constraints or home environment conditions. To address this, we propose the first clinician-led paradigm for real-time generation of personalized rehabilitation software powered by large language models (LLMs). Our system parses natural-language prescription instructions via LLMs and orchestrates smartphone-based sensors to autonomously generate exercises, guide execution, and monitor movement performance. It achieves 99.78% instruction translation accuracy, 88.4% performance monitoring accuracy, and increases personalized intervention coverage by 45% over template-based approaches. Safety assessments by 90% of clinicians met regulatory thresholds, and 75% expressed willingness to adopt the system. This work overcomes the rigidity of conventional parametric interventions, substantially enhancing clinical flexibility, personalization fidelity, and remote management capability in digital rehabilitation.
Application Category
📝 Abstract
Digital health interventions are increasingly used in physical and occupational therapy to deliver home exercise programs via sensor equipped devices such as smartphones, enabling remote monitoring of adherence and performance. However, digital interventions are typically programmed as software before clinical encounters as libraries of parametrized exercise modules targeting broad patient populations. At the point of care, clinicians can only select modules and adjust a narrow set of parameters like repetitions, so patient specific needs that emerge during encounters, such as distinct movement limitations, and home environments, are rarely reflected in the software. We evaluated a digital intervention paradigm that uses large language models (LLMs) to translate clinicians' exercise prescriptions into intervention software. In a prospective single arm feasibility study with 20 licensed physical and occupational therapists and a standardized patient, clinicians created 40 individualized upper extremity programs (398 instructions) that were automatically translated into executable software. Our results show a 45% increase in the proportion of personalized prescriptions that can be implemented as software compared with a template based benchmark, with unanimous consensus among therapists on ease of use. The LLM generated software correctly delivered 99.78% (397/398) of instructions as prescribed and monitored performance with 88.4% (352/398) accuracy, with 90% (18/20) of therapists judged it safe to interact with patients, and 75% (15/20) expressed willingness to adopt it. To our knowledge, this is the first prospective evaluation of clinician directed intervention software generation with LLMs in healthcare, demonstrating feasibility and motivating larger trials to assess clinical effectiveness and safety in real patient populations.
Problem
Research questions and friction points this paper is trying to address.
Digital therapy software lacks personalization for individual patient needs
Clinicians cannot customize exercises during patient encounters effectively
Template-based systems limit adaptation to specific movement limitations
Innovation
Methods, ideas, or system contributions that make the work stand out.
LLMs translate clinician prescriptions into software
Automated generation of personalized therapy interventions
Enables real-time adaptation to patient-specific needs
🔎 Similar Papers
No similar papers found.
